Biology subjects

Borst, J. P.

Publications and source records attributed to Borst, J. P..

7 recordsLinked to original sources

bioRxiv2025

Hierarchical Semi-Markov Smooth Models of Latent Neural States

Hidden (semi-) Markov Models (HsMMs) are increasingly being used to segment neurophysiological signals into sequences of latent cognitive processes. The idea: different processes will leave distinct traces in trial-level recordings of (multivariate) neuro-physiological signals. Markov models, equipped with an emission model of these traces and a latent process model describing the progression through the different latent processes involved in a task, can then be used to infer the most likely process for any time-point and trial. However, the currently used HsMMs remain limited in two important ways. First, they cannot account for subject-level heterogeneity in the latent and emission process. Instead, a single group-level model is assumed to explain the entire data. Second, they cannot account for the potentially non-linear effects of experimental covariates on the latent and emission process. To address these problems, we present a modeling framework in which the HsMM parameters of the emission and latent process are replaced with mixed additive models, including smooth functions of experimental covariates and random effects. We derive all necessary quantities for empirical Bayes and fully Bayesian inference for all parameters and provide a Python implementation of all estimation algorithms. To demonstrate the advantages offered by this framework, we apply such a multi-level model to an existing lexical decision dataset. We show that, even in such a simple task, not all subjects rely on the same processes equally and that at least two semi-Markov states, previously believed to reflect distinct processes, might actually relate to the same cognitive process.

neuroscience↗

bioRxiv2025

Sustained alpha oscillations serve attentional prioritization in working memory, not maintenance

Recent theory on the neural basis of working memory (WM) has attributed an important role to "activity-silent" or -quiescent mechanisms, suggesting that sustained neural activity might not be essential in the retention of information. This idea has been challenged by reports of ongoing neural activity in the alpha band during WM maintenance, however. The precise role of these alpha oscillations is unclear: Do they reflect attentional prioritization of stored information, or do they serve as a general maintenance mechanism, for instance to periodically refresh synaptic traces? To address this, we designed a visual WM task involving two memory items, one of which was prioritized by being tested first for recall. The task included both short (1 s) and a long (3 s) delay intervals between encoding and retrieval. The long delay condition allowed us to test whether the alpha-based decoding effects persist beyond the early delay period, thereby putting accounts that attribute alpha activity to generic maintenance processes to the test. Time-resolved decoding analyses revealed that both tested-first and tested-second items were initially decodable following stimulus presentation. However, only the tested-first item exhibited sustained decodability throughout the delay, particularly in the long delay condition, where it transitioned into a stable coding scheme. This prolonged representation was selectively supported by induced alpha power, which reliably tracked the prioritized tested-first item, but not the deprioritized tested-second one. Impulse-based decoding further confirmed this asymmetry, showing a selective increase in readout for the tested-second item only when it became immediately task-relevant. Together, these findings suggest that sustained alpha-band activity primarily reflects attentional prioritization, rather than general memory maintenance. Unattended, deprioritized items appear to transition into an activity-quiescent state, consistent with models of synaptic storage in WM.

neuroscience↗

bioRxiv2025

Decision-making components and times revealed by the single-trial electro-encephalogram

Decision-making stems from a sequence of information processing steps between the onset of the stimulus and the response. Despite extensive research, uncertainty remains about the actual cognitive sequence involved that leads to the reaction time. Using the hidden multivariate pattern method we modeled the single-trial electroencephalogram of participants performing a decision task as a sequence of an unknown number of events estimated as trial-recurrent, time-varying, stable topographies. We provide evidence for five events occurring during participants decision making and capturing encoding, attention orientation, decision and motor execution times. This interpretation is supported by the observation that a targeted manipulation of stimulus intensity yields Pierons law in the interval between encoding and attention orientation, and Fechners law in the interval between attention orientation and decision commitment. The final, decision-related, event is represented in the brain as a ramping signal in parietal areas whose timing, amplitude and build-up predict the participants decision accuracy.

neuroscience↗

bioRxiv2024

Trial-by-trial detection of cognitive events in neural time-series

Measuring the time-course of neural events that make up cognitive processing is crucial to understand the relation between brain and behavior. To this aim, we formulated a method to discover a trial-wise sequence of events in multivariate neural signals such as electro- or magneto-encephalograpic (E/MEG) recordings. This sequence of events is assumed to be represented by multivariate patterns in neural time-series, with the by-trial inter-event intervals following probability distributions. By estimating event-specific multivariate patterns, and between-event time interval distributions, the method allows to recover the by-trial location of brain responses. We demonstrate the properties and robustness of this hidden multivariate pattern (HMP) method through simulations, including robustness to low signal-to-noise ratio, as typically observed in EEG recordings. The applicability of HMP is illustrated using three previously published datasets. We show how HMP provides, for any experiment or condition, an estimate of the number of events, the sensors contributing to each event (e.g. EEG scalp topography), and the intervals between each event. Traditional exploration of tasks cognitive structures and electrophysiological analyses can thus be enhanced by HMP estimates.

neuroscience↗

bioRxiv2024

Featural and spatial interference with functionally active and passive items in working memory

Functionally active and passive states in working memory have been related to different neural mechanisms. Memoranda in active states might be maintained by persistent neural firing, whereas memoranda in passive states might be maintained through short-term synaptic plasticity. We reasoned that this might make these items differentially susceptible to interference during maintenance, in particular that passively maintained items might be more robust. To test this hypothesis, we gave our participants a working memory task in which one item was prioritised (active) by always probing it first, while the other item was deprioritised (passive) by always probing it second. In two experiments, on half the trials, we presented an interfering task during memory maintenance, in which the stimuli matched either the feature dimension of the memory items (colour or orientation), or their spatial location. Whether the interfering task appeared on a given trial was unpredictable. In a third experiment where participants were given prior knowledge of the interference condition, and finally in a fourth experiment we used a reward-based prioritisation cue. Across experiments, we found that both active and passive memory items were affected by interference to a similar extent, with overall performance being closely matched in all experiments. We further investigated precision and probability of target response parameters from the standard mixture model, which also showed no differences between states. We conclude that active and passive items, although potentially stored in different neuronal states, do not show differential susceptibility to interference. Public significance statementThe ability to briefly remember information is critical to human cognition. Our so-called working memory is nevertheless rather limited, able to hold only a few items at any one time, and prone to forgetting when we are briefly distracted. Yet, there is reason to believe that not all information in working memory is equally vulnerable. Items that are more passively stored, because they will only be required after some time, might be more resilient to interference. Items that are stored actively, for more immediate recall, might be more easily disrupted. Here, we investigated the effect of an interference task on the retention of both active and passive items in working memory. Our results showed that active and passive items are equally affected by interference, suggesting that resilience in working memory does not depend on the functional state of the items therein.

neuroscience↗

bioRxiv2020

Decoding Study-Independent Mind-Wandering from EEG using Convolutional Neural Networks

ObjectiveMind-wandering is a mental phenomenon where the internal thought process disengages from the external environment periodically. In the current study, we trained EEG classifiers using convolutional neural networks (CNN) to track mind-wandering across studies. ApproachWe transformed the input from raw EEG to band-frequency information (power), singletrial ERP (stERP) patterns, and connectivity matrices between channels (based on inter-site phase clustering, ISPC). We trained CNN models for each input type from each EEG channel as the input model for the meta-learner. To verify the generalizability, we used leave-N-participant-out crossvalidations (N=6) and tested the meta-learner on the data from an independent study for across-study predictions. Main resultsThe current results show limited generalizability across participants and tasks. Nevertheless, our meta-learned trained with the stERPs performed the best among the state-of-theart neural networks. The mapping of each input model to the output of the meta-learner indicates the importance of each EEG channel. SignificanceOur study makes the first attempt to train study-independent mind-wandering classifiers. The results indicate that this remains challenging. The current proposed stacking neural network design allows an easy inspection of channel importance and feature maps.

neuroscience↗

bioRxiv2019

A functional spiking-neuron model of activity-silent working memory in humans based on calcium-mediated short-term synaptic plasticity

In this paper, we present a functional spiking-neuron model of human working memory (WM). This model combines neural firing for encoding of information with activity-silent maintenance. While it used to be widely assumed that information in WM is maintained through persistent recurrent activity, recent studies have shown that information can be maintained without persistent firing; instead, information can be stored in activity-silent states. A candidate mechanism underlying this type of storage is short-term synaptic plasticity (STSP), by which the strength of connections between neurons rapidly changes to encode new information. To demonstrate that STSP can lead to functional behavior, we integrated STSP by means of calcium-mediated synaptic facilitation in a large-scale spiking-neuron model and added a decision mechanism. The model was used to simulate a recent study that measured behavior and EEG activity of participants in three delayed-response tasks. In these tasks, one or two visual gratings had to be maintained in WM, and compared to subsequent probes. The original study demonstrated that WM contents and its priority status could be decoded from neural activity elicited by a task-irrelevant stimulus displayed during the activity-silent maintenance period. In support of our model, we show that it can perform these tasks, and that both its behavior as well as its neural representations are in agreement with the human data. We conclude that information in WM can be effectively maintained in activity-silent states by means of calcium-mediated STSP. Author SummaryMentally maintaining information for short periods of time in working memory is crucial for human adaptive behavior. It was recently shown that the human brain does not only store information through neural firing - as was widely believed - but also maintains information in activity-silent states. Here, we present a detailed neural model of how this could happen in our brain through short-term synaptic plasticity: rapidly adapting the connection strengths between neurons in response to incoming information. By reactivating the adapted network, the stored information can be read out later. We show that our model can perform three working memory tasks as accurately as human participants can, while using similar mental representations. We conclude that our model is a plausible and effective neural implementation of human working memory.

neuroscience↗